What is the syntax-semantics interface?

The syntax-semantics interface simply refers to how syntax and semantics talk to each other.

Recall that syntax is the part that’s responsible for the structure of language while semantics is the part that’s responsible for the meaning of language.

Think of syntax and semantics are two black boxes within the bigger black box that is human language.

You can think of the syntax-semantic interface as the communication wires between these two black boxes that transfer information between them.

In simple sentences, there is a one-to-one correspondence between each element of structure and each element of meaning.

But in many other sentences, structure and meaning come apart.

Case study 1:
Scope ambiguity

Scope ambiguities are a textbook example of a mismatch between structure and meaning.

Take a minute to think about the meaning of the sentence in (1).


John doesn’t think he’s smart.

Does this sentence mean:

(a) John does NOT THINK: “I’m smart”?

Or does it mean:

(b) does it mean that John THINKS: “I’m NOT smart”?

If (b) is the correct reading (or the most prominent one), why does not (n’t) come before think in (1)?

Linguists would say that in (a), NOT takes scope over THINK. Since this meaning matches the form of the sentence in (1), this is sometimes referred to as the surface scope reading.

Linguists would say that in (b), THINK takes scope over NOT. Since the meaning is the opposite of what you might expect from the form of the sentence, this is sometimes referred to as the inverse scope reading.

The phenomenon where one sentence has more than one scope reading is referred to as scope ambiguity.

In one of my research projects, I investigate why Mandarin sentences with reciprocals like bǐcǐ and hùxiāng exhibit scope ambiguity.

Case study 2:
Argument structure

Another active research area in the syntax-semantics interface is argument structure.

You can think of arguments as the participants in a sentence. In (2) and (3), man and dog are arguments of the verb bites.


Man bites dog.


Dog bites man.

Syntax assigns grammatical roles like subject and object to arguments.

In some languages like Japanese, subjects and objects are marked by different word endings (-ga and –o).

In other languages like an English, subjects and objects are distinguished by word order. For example, John is the subject of (4) but the clothes is the subject of (5).


John washed the clothes.


The clothes were washed by John.

The two sentences (4) and (5) have pretty much the same meaning.

In both sentences, John is the agent of washing and the clothes are the patient or theme of washing.

Semantics assigns thematic roles like agent and theme to arguments.

But the mapping between the syntax and semantics is not transparently one-to-one.

In (4), syntax assigns the grammatical role of subject to John while semantics assigns the thematic role of agent to John.

In (5), syntax assigns the grammatical role of subject to the clothes, but this time semantics assigns the thematic role of theme to the clothes.

How then are grammatical roles like subject and object mapped onto thematic roles like agent and theme?

Perhaps you may be thinking: isn’t it obvious from the context? And indeed, in many simple cases, listeners could guess the correct interpretation of a sentence from context.

But in other cases, things get more complex.

In my research on Mandarin resultatives, I consider sentences like (6) which has three possible interpretations.


Táotáo zhuī-lèi-le Yǒuyǒu.
Taotao chase-tired-PFV Youyou
(a) ‘Taotao chased Youyou and Youyou got tired.’
(b) ‘Taotao chased Youyou and Taotao got tired.’
(c) ‘Youyou chased Taotao and Youyou got tired.’
(d) Not: ‘Youyou chased Taotao and Taotao got tired.’

(Li 1995)

The subject Táotáo can be interpreted as the agent or theme of chasing, as well as the entity that got tired or not.

One possible hypothesis is that when Mandarin speakers encounter a sentence like (6), they make a good guess as to who does what to whom based on the context.

But this hypothesis cannot be correct because speakers never accept the fourth logical possibility in (6d).

In my research, I explain why sentences with Mandarin resultatives have such flexible interpretations.


The term “syntax-semantics interface” refers to how syntax (the part of language responsible for structure) and semantics (the part responsible for meaning) talk to each other.

Understanding how the syntax-semantics interface works is crucial for understanding how language works.


I hope you found this post helpful, but if you have any questions you’d like to see answered, please feel free to contact me!